Ecological Applications of Stand Density Indices chaired by Mark Ducey

Density-Size Relationships for Sitka Spruce and Douglas-Fir in Britain and Canada and Linkage to Light Capture

P. Comeau — Tue, 23 Jun 2009 11:30:00 PDT

Density-size relationships have potential value in guiding thinnings and in the development of shelterwood and selection systems. Reineckes stand density index (SDI) is used by some as a measure of site occupancy and the degree to which site resources are fully captured. While a few studies indicate that SDI may be more strongly related to light capture than basal area or other stand attributes, linkages between light capture and density-size relationships need further exploration. A better understanding of how density-size relationships are influenced by environment and how changes in SDI are linked to changes in resource utilization by stands is also of ecological interest. Sitka spruce and Douglas fir from western North America have been widely planted in Great Britain. Sitka spruce comprises 49% and Douglas-fir represents 3% of the total conifer forest cover in Great Britain. During 2008 I initiated a study to: 1) Develop and evaluate density-size relationships for Sitka spruce and Douglas-fir stands in the UK (using data collected by the Forestry Commission); 2) Compare these to density-size relationships for Sitka spruce and Douglas-fir stands in British Columbia (using government and industry PSP data); and, 3) Examine relationships between understory light levels (measured using hemispherical photography) and SDI (or other stand level measurements) in Sitka spruce stands and Douglas-fir stands. Preliminary results indicate that slopes for density-size relationships are steeper than -1.6 for both species in Britain and that site quality (yield class estimated from site index) has a small but significant influence on the position of the density-size line (but not the slope). For both species, basal area performs as well as or better than SDI for estimating light capture.

Canopy Cover Prediction From Stand Density Attributes: Stocking, Crown Width, and Overlap Functions

Andrew Gray et al. — Tue, 23 Jun 2009 11:50:00 PDT

The goals for a wide range of forest management objectives are often stated in terms of the amount and layering of canopy cover. However, measuring canopy cover is labor intensive and different techniques provide widely different estimates. Several approaches have been developed to predict cover from common tree or stand-level density attributes, with varying results. This study used line-intercept measured tree cover from 1,424 Forest Inventory and Analysis (FIA) plots across Oregon to build predictive models from estimates of tree stocking, crown width, and other stand attributes (mean diameter, stand height, SDI, etc.). A variety of adjustments were applied to adjust for tree social status and account for tree crown overlap. Stocking was a better predictor of cover than crown width, although much of the error in the latter was due to estimates of crown overlap. The random crown overlap function that is standard in the Forest Vegetation Simulator (FVS) resulted in biased predictions in mesic forest types, but not in dry forest types. New model predictions based on stocking for mesic forest types were within 15 percent of measured cover for > 82% of the observations. Although there are some additional options for improving estimated cover from tree and stand attributes, ground-based measurements will probably be required for precise estimates.

Utilizing Quantile Regression to Develop a Stand Density Index for Mixed Species Stands From FIA Data

R. Knapp et al. — Tue, 23 Jun 2009 13:30:00 PDT

Identifying relative stand density values is important for accurate prediction of net biomass and carbon accumulation, for estimating growth rate and mortality risks of individual trees, and managing stands for multiple goods and services. Developing stand density indices for mixed species stands has been particularly challenging. Utilizing data from the USDA Forest Inventory and Analysis database, we developed a mixed species stand density index for the northeastern United States. This additive index utilizes quantile regression to identify maximum stand density values for mixed species stands. A quantile selection technique was used to ensure that single-species maximum density values calculated using the new index are in agreement with currently accepted or implied maximum values contained in other silvicultural guidelines. This technique has broad application and offers a robust way for dealing with the complex structures of mixed species stands.

Influence of Relative Density and Composition on Growth Rates in Boreal Mixedwoods

V. Reyes-Hernandez et al. — Tue, 23 Jun 2009 13:50:00 PDT

Self-thinning or density-induced mortality is the consequence of competition between individuals of a stand and takes place when resources for tree development become insufficient for individuals needs or for the stand as a whole. Since density by itself is not a good indicator of competition between component species, other expressions -like Reineke’s stand density index (SDI) - have been proposed as suitable indicators of site occupancy and competition. Specific objectives of this research are: a) to develop density - size relationships and relative density indexes for aspen and white spruce mixtures and, b) to investigate the effect of composition and site quality on maximum density - size relationships and stand growth rates. We are using data from permanent sample plots for trembling aspen ñ white spruce stands from western Canada (Alberta, Saskatchewan and Manitoba) to examine the usefulness of density - size relationships in mixed and pure stands of these species. Modeling of these relationships is being done with data that cover a wide range of ages, densities, species compositions and site conditions. Multiple linear and non-linear regression are used to examine the relationship among maximum density and growth rates with quadratic mean diameter, composition, site quality and other site quality indicators. Our data suggest that a number of factors may influence the intercept value of the self-thinning lines but that the slope of these lines remains nearly constant. Preliminary results also show that the upper limit of density is related to quadratic mean diameter of the stand, composition (percentage of total basal area occupied by aspen) and site index. Our results will provide insight into the effects of composition and site quality on maximum density- size relations and stand growth rates, and into the potential application of these relationships in predicting mixedwood stand development.

Stand Density in South Florida Tropical Forests: Implications for the Function and Management of Everglades Tree Islands

M. Ross et al. — Tue, 23 Jun 2009 14:10:00 PDT

Within the continental US, the broadleaved forests of south Florida are exceptional in the abundance and diversity of tree species of tropical origin. Dry tropical forests are regionally most extensive in the upper Florida Keys, but are also represented on the mainland as fragments on limestone rocklands, and as “tree islands” embedded in the Everglades marsh. The exposed Everglades tree islands have a history of human use reaching back thousands of years, and are subject to frequent disturbance from tropical storms and hurricanes. They are sensitive to the hydrology of the surrounding marsh, which can lead to gradual changes in species composition or stand structure, or to the sudden loss of the woody component entirely, especially when low water tables are precursors to damaging fires. Tree islands serve as local hotspots of biodiversity, and as concentrators of phosphorus in a landscape defined by P-limitation. The mechanisms by which P reaches the tree islands and is sequestered there are complex and not completely understood, but may depend in part on transpiration and resupply of water from the adjacent wetlands. Since transpiration is a direct function of the transpiring leaf surface, which itself is expected to vary with stocking, we examined the relationship between leaf area index and stand density in 16 Everglades tree islands. To determine maximum stocking levels for such forests, we also calculated stand density for tropical forests throughout south Florida, using a protocol modified slightly from Woodard et al. 2003. Our results suggest that (1) stand density in many Everglades tree islands is well below the expressed potential of similar tropical assemblages, (2) low site occupancy may prevent such under-stocked forests from performing several ecosystem functions, and (3) stand density can serve as an effective metric of forest condition for management or restoration purposes.

Can Live Tree Size-Density Relationships Provide A Mechanism for Predicting Down and Dead Tree Resources?

C. Woodall et al. — Tue, 23 Jun 2009 14:30:00 PDT

Live tree size-density relationships in forests have long provided a framework for understanding stand dynamics. There has been little examination of the relationship between the size-density attributes of live and standing/down dead trees (e.g., number and mean tree size per unit area, such information could help in large-scale efforts to estimate dead wood resources. The goal of this study was to examine the relationship between standing live, standing dead, and downed dead trees in the context of size-density attributes using a national inventory of forests. Our results indicated that from the lowest to the highest live tree relative stand density, the mean biomass/ha of live trees increased by more than 2,000 percent while the mean biomass/ha of standing dead and downed dead trees increased 295 and 75 percent, respectively. Correlations between downed dead wood and stand/site attributes reached their highest level (r > 0.60) when a stand’s relative density exceeded 80 percent. We propose a model for highly stocked stands whereby downed and dead wood biomass may be predicted based on live/dead tree size-density attributes, stand age, and climatic factors. We also provide an alternative model for moderate / low stocked stands whereby potential maximum live biomass may serve as a limit to dead wood resources with stochastic events (e.g., wind/mortality disturbances) as high-impact variables. Overall, the size-density attributes of live/dead trees may help guide the estimation of downed and dead wood attributes in forests.